International Standards Organization (“ISO”) Standard shipping containers are used worldwide for moving goods of virtually all types, including raw materials, manufactured goods and components, vehicles, dry and refrigerated foodstuffs, and hazardous materials. All ISO Standard shipping containers have a common width, 8′-0″, and a fixed set of modular lengths: 10′-0″, 20′-0″, 30′40″, and 40′-0″. The vast majority of shipping containers in use are either 20′ or 40′ long. A “twenty-foot equivalent unit”, or TEU, represents a single 20′ ISO container. A 40′ ISO container is two TEUs. These ISO containers come in a number of technical variants, including refrigerated containers for perishables, tanks for handling fluids, flat racks and bulkhead racks for handling oversized cargo, car containers, livestock containers, and others.
In addition to the standard ISO container sizes, specialty containers based on the same technology have been put into service in the United States for domestic transport via truck or rail. These containers currently come in 45′, 48′, and 53′ lengths, but have internal framing allowing them to be handled by the same equipment used for 40′ ISO containers. Use of a common width and modular lengths has allowed the creation of an extremely efficient and cost-effective system of global freight transportation. The development of an ISO Standard for container dimensions has allowed a globalization of the machinery, systems, techniques, and infrastructure used to handle the containers.
Container transport by sea is largely done in vessels specifically designed to accommodate, protect, and rapidly move standard shipping containers. Most such vessels are “cellular” in design, with most containers stored below-decks among vertical stanchions that provide container and vessel stability. Vessels come in a wide range of sizes and operational configurations depending on the specific needs of the markets they serve. Vessel deployment changes rapidly over time in response to dynamic market changes.
While shipping containers have been standardized, the vehicles for carrying them have not. Waterborne vessels, trucks, and trains are designed within the context of individual national transport and safety regulations and in response to local and regional market forces. While there is some commonality in dimensions driven by the standardization of container dimensions, overall shapes and dimensions show considerable variability.
Port container facilities provide infrastructure, machinery, and other resources for transferring containers between water-borne and land-borne transport modes. FIG. 1, discussed in detail below, shows the key operational elements of a generic port container facility.
One of the difficulties in designing port facilities is that each site is different. In particular, the geography varies greatly. The specific design of each port container facility must reflect both the standardized elements of container handling and many site-specific geometric and design constraints, along with market-specific logistical and regulatory demands. As such, each port container terminal is unique in its configuration, capacity, productivity, efficiency, safety, and environmental footprint.
To overcome site variability while still providing efficiency in a system that serves a standardized freight module, a variety of common sub-system elements and machines have evolved, ready for selective deployment in each facility as local conditions require.
For example, containerized shipping is inherently flexible because of the modularity of the shipping container. Containerization allows the mixing of a very wide variety of freight types within a single facility or within a single vessel. While the shipping containers are flexible in their deployment, handling, storage, and use, it is still true that each freight unit being moved within a container retains its own logistical needs. All systems involved in container handling must be ready to support the logistical needs of freight beneficial owners and regulatory agencies, while still supporting high productivity and maximum protection for the freight and workers.
Because of the variability of transport vehicles, the variability of port infrastructure, and the variable deployment of container handling equipment, port container operations have traditionally relied largely on maimed equipment rather than robotics. The complex interplay of machines has required “human eyes” to ensure the safety of both transport and maritime workers and machines. The demanding physical environment outdoors on the waterfront has made it very difficult to develop effective and reliable instrumentation to substitute for human capabilities in port facilities.
Some activities in the port terminal require close interaction between workers and containers or machines. Container securing devices known as “inter-box connectors” (IBCs) or “cones” are used to hold containers in place on waterborne vessels, and can only be effectively handled by workers. Refrigerated containers (“reefers”) must be connected and disconnected against shore power outlets by human workers, and most terminals rely on workers to check the status of reefers while in storage. Container handling equipment frequently needs close attention by mechanics for routine diagnosis, maintenance, and repairs, as well as for swapping specialized cargo-handling hardware.
The complex, variable, and demanding environment of the port container terminal has resisted the development and deployment of automation systems. To date, only a handful of terminals make use of robotic container handling equipment that operates without direct human tactical control. So far, only three types of container handling equipment have been deployed in a robotic configuration: 1) Automated Stacking Cranes (ASCs); 2) Automated Guided Vehicles (AGVs); and 3) Automated Van Carriers (AVCs), each of which is described briefly in the following paragraphs.
AVCs are diesel-electric or diesel-hydraulic machines used to store, retrieve, and transport containers within a fully-robotic container yard designed for AVC-configured storage, as shown in FIG. 2. Each AVC can pick, set, or transport a single container to or from storage stacks that are one container wide and up to three containers high. AVCs interact with manned quay cranes on the waterside edge of the terminal via a “grounded buffer” in the crane's working envelope. See FIG. 3. They interact with manned street trucks on the landside edge of the terminal by having a human operator take control of the AVC and operate it using remote-control to pick or set a container while compensating for variability in truck placement or configuration. See FIG. 4. AVCs have not been developed to interact with on-dock intermodal rail yard equipment; this movement must be done via manned trucks or with manned van carriers.
By far the most common ASCs are electric rail-mounted gantry cranes used to store, retrieve, and rehandle containers within a high-density storage block, typically six to ten container stacks wide and three to five containers high. See FIGS. 5A and 5B. ASCs use their rail gantry drives to transport containers to either end of their storage blocks, which are oriented with their rails perpendicular to the wharf. At the waterside edge of the container yard, ASCs have been used to interact with either AGVs or with Manned Van Carriers (MVCs). At the landside edge of the container yard, ASCs have been used to interact with either street trucks or MVCs.
AGVs are diesel-electric transport units capable of transporting a single container from point to point within a fully-robotic operating zone. See FIGS. 6A through 6E. The AGV cannot pick or set a container as an AVC can, but must rely on direct service by an overhead crane, typically either the ASC in the container yard or the quay crane on the wharf.
In all cases to date, interactions between ASCs and manned machines, whether they are MVCs, street trucks, or yard trucks, require direct intervention by a human operator, working the ASC in remote-control mode. The remote-control operators rely on cameras, lasers, and other instruments to achieve proper container alignment and maintain operational safety. The ASC/manned interface buffer is a complex area, with many safety interlocks intended to protect human workers, with “fail-safe” design that prohibits operation if the safety interlocks are determined to be dysfunctional for either mechanical or environmental reasons. The ASC/manned interface buffer must be well-lit for safety under all operating conditions, contributing to night-time light pollution. See FIG. 7.
ASCs have been deployed in two general configurations. The “common rail” configuration has one or more ASCs of common rail gauge working over the same storage block, riding on the same pair of gantry rails. The “nested rail” configuration has one or more ASCs of two rail gauges working over the same storage block, riding on separate parallel sets of gantry rails, with a smaller ASC able to pass beneath and within the interior envelope of a larger ASC.
In the “common rail” ASC configuration, the ASCs cannot pass one another. It is generally very difficult or impossible for two such ASCs to share work at the end interface zones: one ASC must be dedicated to serve each end of the storage block. In general, the landside ASC cannot be used to augment productivity on the waterside end of the block, and vice versa. See FIG. 8.
In the “nested rail” ASC configuration, the larger and smaller ASCs can pass one another, albeit with limitations designed to prevent loads being carried by the larger ASC from colliding with the smaller ASC during gantry movement. The two ASCs can be used together to augment productivity at one end of the block or the other, but safety limitations prevent taking maximum advantage of this capability, especially at the landside end of the block where manned transport vehicles are being served. See FIGS. 9 and 10.
The “common rail” ASC configuration is more compact because there are only two rails per block instead of four, no empty space dedicated to allowing suspended loads to safely pass, and only one crane power supply cable runway instead of two. This configuration has high storage density and high terminal capacity, but less productive flexibility than the “nested rail” configuration. The “nested rail” ASC configuration takes up more space, supporting lower storage density and terminal capacity while providing greater productive flexibility. There is no settled industry paradigm identifying either configuration as the “best practice”.
The need to separate manned truck or manned van carrier operations from robotic AGV operations has caused virtually all current ASC-based terminals to: 1) run ASC blocks perpendicular to the wharf; 2) use the container yard ASC blocks as a barrier between manned and robotic transport systems. This paradigm works well when there is sufficient land depth perpendicular to the wharf to allow for a proper balance between operational productivity and static storage capacity. For terminals with a limited land depth, this paradigm limits the terminal's storage capacity and can make automation of the facility unfeasible.
Most port container terminals are managed with the support of a “Terminal Operating System” (TOS). The TOS generally encompasses: 1) a sophisticated database system; 2) one or more operational planning tools for organizing terminal activities; 3) commercial communication utilities for coordinating internal and external activities; 4) financial and billing systems supporting interaction between the terminal operator, customers, and regulators; 5) graphical user interfaces that ease data input, output, and manipulation; and 6) labor utilization recording and reporting software in support of financial activities. The TOS is an essential management tool, but most TOS packages were not designed to support or direct robotic automation of container handling. TOS packages are available from a number of vendors including Navis, Cosmos, Tideworks Technology, Total Soft Bank, Realtime Business Solutions, and Embarcadero Systems.
A typical TOS is depicted diagrammatically in FIG. 28. A typical TOS includes internal systems for planning, inventory management, and job control. It includes internal messaging systems that allow transfer of information to and from container yard, vessel, and rail operating elements inside the terminal, as well as external messaging for transfer of information to and from a range of commercial and governmental entities outside the terminal, via Electronic Data Interchange (EDI) or through customer-specific Web Portals.
Each existing and currently-planned automated container terminal has some form of “Equipment Control System” (ECS). Although there is considerable variability in architecture, and in the relationship with the TOS, the ECS generally encompasses: 1) tracking of robotic movement; 2) directed control of robotic movement; 3) processing of remote-control sensor and operating signals; 4) processing of detection sensor information from the manned interface areas; 5) simulation or emulation algorithms for predicting current or future robotic duty cycle performance; 6) sophisticated artificial intelligence for assigning robotic equipment among competing demands based on dynamic prioritization logic; 7) sophisticated artificial intelligence for allocating container storage space while balancing current and future robotic performance for storage and retrieval; and 8) communication and coordination with the TOS and with robotic equipment. ECS packages are available from a number of vendors including Gottwald, TBA, ABB, Hamburg Port Consulting, Informs, and Navis.
A typical ECS is depicted diagrammatically in Figure N. It includes interfaces with the TOS's in-terminal messaging systems, using data from the TOS to dispatch automated equipment to serve jobs in the correct order and with maximum efficiency. The ECS interfaces directly with the hardware and software systems in each automated or manned piece of equipment, pulling information about status, performance, speed, and other factors, and transmitting movement, routing, container handling, and other instructions in the appropriate order.
The TOS and ECS must work together with negligible human intervention for successful operation of an automated terminal. The scopes and capabilities of the TOS and the ECS in the automated realm vary considerably from one facility to the next, and there is no settled paradigm providing a “best of class” configuration for balanced, coordinated activity.
The primary difficulty facing TOS and ECS joint development is the blending of two sets of activities: 1) fully-robotic, predictable, plannable container movements that do not need to protect workers; 2) manned, unpredictable, unplannable container movements that involve substantial robotic safeguards for workers. Theoretically, the TOS has data describing future logistical demand, but these data are frequently imperfect and are never at the level of “tactical granularity” required for the ECS to perfectly understand future demand. The artificial intelligences embodied in the ECS cannot create the “perfect solution” for every array of competing tasks and activities because they cannot predict future manned operating demand or performance. Each ECS must make “best-guess” choices based on rough simulation and emulation calculations, and have some ability to correct or change instructions as the operation evolves. There is no settled paradigm for establishing or applying such corrections.
In summary: 1) Containers are standardized but transport and terminals are not. 2) Terminal operations are variable and resistant to full automation. 3) Some terminal equipment and systems have been the focus of robotization, resulting in a small range of robotic equipment and semi-automated terminals. 4) All existing automated terminals include a complex blend of manned and robotic operations. 5) The blending of manned and robotic operations leads to unpredictability in the demand on the robotic system and complexity in the relationship between workers and robots. 6) Terminal and equipment control systems cannot fully optimize equipment and resource allocation because of the inherent randomness of mixed manned/robotic operation. 7) The need to separate manned and robotic in-terminal transport has limited the applicability of ASC-based systems to terminals with greater land width perpendicular to the wharf.
The current automation paradigms could be improved upon as follows. Eliminate direct interaction between robotic yard cranes and manned transport equipment. Allow required manned activities to take place safely without interfering with the performance of the robotic system. Allow automation of landside container transport, including transport to and from in-terminal rail facilities. Allow full robotization of container storage and retrieval under centralized and optimized TOS/ECS control. Combine the high storage density and capacity associated with “common rail” ASC systems with the high productivity and flexibility of “nested rail” ASC systems. Allow seamless, balanced sharing of waterside and landside operating demands and ASC allocations. Allow semi-automation of in-terminal rail train load and discharge operations. Largely eliminate the impact of random manned container tasks in the centralized optimization of robotic equipment assignment within the ECS. Eliminate the need for fail-safe robotic/manned interfaces between yard cranes and landside transport equipment. Allow terminals with limited land width to take advantage of ASC-based automation.